[1.12.x] Kerbal Atomics: fancy nuclear engines! (August 18, 2024)

[RSX2400]

On 6/1/2020 at 12:59 PM, Nertea said:

Would like to see some screenshots of the ship in question.

The large nuclear engine ship I've had the most success with is this guy:

It's a single Liberator with 6 of the large graphine radiators. It doesn't overheat, thankfully.

If I use the Emancipator in place of the Liberator, the engine will overheat very quickly.

The most vexing engines are the nuclear aerospikes. I mostly use the close core engine, below, but I can only run the engine at full power for a couple of minutes before it begins to overheat. This ship uses 8 of the smaller extendable graphine radiators (it also has one of the 2 meter fission reactors and an a 2 meter ISRU embedded under the tanks). I can keep the heat stable of I run it at 65% throttle or lower.

The Stubber motor I've had a lot more success with. I run a cluster of 6 of those engines with the large graphine radiators for around 15 minutes before I start to have heat issues with them.

Since my last post, I've unlocked the high-end fission reactors and the Colossus MHD from your other mods, and those have become my go-to long haul propulsion. I've built a very solid propulsion bus with them for a planned Jool mission. I've attached the below image just to show off.

Also, do you have any plans for tackling fusion power and propulsion. I tried KSP Interstellar over the weekend, but that mod has got a ton of stuff to unpack...

[Nertea]

To balance things they generate a lot of heat, long burns on the order of 15 minutes will definitely need reduced throttle. Might revisit this, I am working on another mod to redo the thermal system overall and it will help these kind of things: 

You could also check out my generic WIP thread for some things you might find interesting.

Edited June 3, 2020 by Nertea

[panarchist]

On 6/3/2020 at 10:27 AM, Nertea said:

To balance things they generate a lot of heat, long burns on the order of 15 minutes will definitely need reduced throttle. Might revisit this, I am working on another mod to redo the thermal system overall and it will help these kind of things: You could also check out my generic WIP thread for some things you might find interesting.

This is one of the reasons why NASA has so many LANTR concept pitches - in addition to greatly increasing the thrust of an NTR, the LOX carries away a lot more heat. Any NTR or other nuclear engine type is going to have much better thermal performance as propellant mass flow increases. (the side effects are increased thrust and decreased Isp) Ideally, any NTR with LOX augmentation would be a lot more prone to failure and more expensive to repair (for the repair mods out there) since oxygen makes the engine environment far more corrosive. There are some good arguments for revisiting the nukes in this mod with an eye towards providing better thermal management and increased thrust with increased fuel use.

That said, I've really like using the mod as-is, and the Liberator is one of my favorite rocket engines despite how hot it gets.

Edited June 5, 2020 by panarchist
Initially said "reduced ΔV" when I meant "reduced Isp"

[tinygrox]

On 4/26/2020 at 6:25 PM, lzq2006 said:

Hello, I am a player from China and I look forward to translate Kerbal Atomics (into Chinese of course); should I also notify the GitHub account "ChrisAdderley" regarding matters of translation?

Dude, I already did.

[Nertea]

5 hours ago, panarchist said:

This is one of the reasons why NASA has so many LANTR concept pitches - in addition to greatly increasing the thrust of an NTR, the LOX carries away a lot more heat. Any NTR or other nuclear engine type is going to have much better thermal performance as propellant mass flow increases. (the side effects are increased thrust and decreased ΔV) Ideally, any NTR with LOX augmentation would be a lot more prone to failure and more expensive to repair (for the repair mods out there) since oxygen makes the engine environment far more corrosive. There are some good arguments for revisiting the nukes in this mod with an eye towards providing better thermal management and increased thrust with increased fuel use.

That said, I've really like using the mod as-is, and the Liberator is one of my favorite rocket engines despite how hot it gets.

That's not really the whole picture, in LANTRs the additional oxygen propellant is injected in the nozzle throat or at the bottom of the actively cooled region, so you are only potentially increasing the efficiency of cooling anything downstream of that. The mass flow through the heating region (the reactor) is fully LH2 and identical in either case. I would be curious as to the trade off of any additional cooling effect - you are increasing the mass flow, but you are also adding a new combustion region that will add heat to the system as well. This might negate the cooling benefits. Injecting the oxygen into the reactor itself is not something that should be contemplated.

Injecting in oxygen when it's gasified but not at very high temperatures would minimize the problem of oxygen corrosion, which is what the Triton system (the one whose papers I've actually read, other might differ). Such as system will actually be less taxing on components than a traditional rocket full flow combustion rocket engine because it limits the amount of hot GOX you're moving around. 

 

6 hours ago, panarchist said:

There are some good arguments for revisiting the nukes in this mod with an eye towards providing better thermal management and increased thrust with increased fuel use.

Can you explain what you mean here? There are some changes that will happen with the new heating stuff I'm developing, always good to get inputs. 

[panarchist]

41 minutes ago, Nertea said:

That's not really the whole picture, in LANTRs the additional oxygen propellant is injected in the nozzle throat or at the bottom of the actively cooled region, so you are only potentially increasing the efficiency of cooling anything downstream of that. The mass flow through the heating region (the reactor) is fully LH2 and identical in either case. I would be curious as to the trade off of any additional cooling effect - you are increasing the mass flow, but you are also adding a new combustion region that will add heat to the system as well. This might negate the cooling benefits. Injecting the oxygen into the reactor itself is not something that should be contemplated.

Injecting in oxygen when it's gasified but not at very high temperatures would minimize the problem of oxygen corrosion, which is what the Triton system (the one whose papers I've actually read, other might differ). Such as system will actually be less taxing on components than a traditional rocket full flow combustion rocket engine because it limits the amount of hot GOX you're moving around. 

 

Can you explain what you mean here? There are some changes that will happen with the new heating stuff I'm developing, always good to get inputs. 

You're right in terms of the reactor - LOX has minimal impact on the reactor, but the combustion chamber and nozzle now have more mass to carry away heat, and additional mass blocking heat which is re-radiated back towards the reactor and the back end of the spacecraft. You may be right that it's offset, I haven't fully crunched the math - but LANTR and other LOX NTRs in "afterburner" mode generate 3x as much thrust using 4x as much mass. So while the operating temperature is higher, total heat is lower since more of it is traveling out with the propellant rather than re-radiating back to the spacecraft. 

re: explain what I meant - there are a lot of propulsion concepts out there where propellant is separate from fuel (almost all the nuke concepts, and a few others which I can't recall at the moment). In many of them, the engines are usually high(er) Isp and low(er) thrust than chemical propulsion, and additional propellant (or another propellant with a higher MMW) is injected to increase the mass leaving the nozzle and improve thrust. In may of those cases, the added propellant acts as a coolant due to a higher specific heat. (an example is a LHyd engine using H2O for thrust) Additionally, there's other ways to reject heat through the use of flash coolers of other types of heat sinks which can be flushed into space (i.e. single use) to reject large amounts of heat other than through radiation. It gets discussed a lot in SF and gaming forums about futuristic space combat, and Ken Burnside's "Attack Vector: Tactical" boardgame actually did the math on it, (I might be able to find that paper) as did Children of a Dead Earth. They came to different conclusions as to the practicality, but STS did the same thing with the ammonia boilers and flash evaporators. Base KSP doesn't do anything with heat sinks, (and they can be somewhat problematic due to mass) but there are plenty of examples of their use in RL space agencies.

Take all that with a grain of salt. I haven't done the math on it in 15 years, and some things may have changed. And I'm over 50 now, so the odds of misremembering are a lot greater. But I helped design and playtest parts of Attack Vector, and used to debate thermodynamics and heat sinks with Winchell Chung, Ken Burnside, Anthony Jackson, and a bunch of defense contractors and space scientists on sfconsim-l and a couple of space physics groups - again, 10-15 years ago. I'm probably better now at generating ideas and pointing people at resources than doing the actual math, but I still remember a few things.

I might be wrong about LANTR LOX augmentation, but thrust power directly relates to thermal energy, so the determining factor would be the propellant ratio and specific heat of the propellants. H2 has 14x the specific heat per mass unit that O2 does, so maybe I have it backwards. 

[RyGuy_McFly]

Hi there, been using your mods for a long time but only just now trying to use nukes. Got my first Nerv in the air today and found out that its reactor doesn't seem to produce electricity. it's not a big deal on my current ship, it's just a small rocket sending a SCANsat probe to Minmus (overkill, I know, but I just wanted to try out this mod), and it has solar panels, but I just wanted to see what kind of power it could produce. I'm not sure if I'm just not understanding it properly, is the reactor in the nuclear engines only required to be on for the engine to work, or is it actually for producing power? The NFE control panel doesn't show the engine as a producer of power, only my solar panels and a different engine with an alternator. I have three basic radiators on it, which can maintain a core temp of 3000K @ 14% power, which apparently should produce 142.6kW of power.

If this is a bug, I can produce screenshots/logs, but I think I might just be misunderstanding the changes made to atomic motors...

[KSPrynk]

18 hours ago, RyGuy_McFly said:

Hi there, been using your mods for a long time but only just now trying to use nukes. Got my first Nerv in the air today and found out that its reactor doesn't seem to produce electricity. 

If this is a bug, I can produce screenshots/logs, but I think I might just be misunderstanding the changes made to atomic motors...

Some of the nuclear thermal rockers are bi-modal (i.e., produce continuous power when not thrusting) and some are not.  This is different from an alternator, which produces power only during thrusting.  Unless you don't like surprises being ruined, I recommend starting a separate sandbox game and checking them all out.

As far as I know, you don't have to turn on a generator in any of the bi-modal NTRs (although there are usually integrated radiators to turn on), like you would for the dedicated nuclear reactor powerplant parts (which come in the Near Future Electrical mod), but I haven't checked to see if I'm a version behind on KA and whether that requirement was recently added.

On the NFE powerplant reactors, you have to start and set the power output level manually.  Just don't set them higher than you need, because Uranium will get consumed and you'll burn it all up on long trips, whether you're using the power or not.  I was reading upthread that the NTRs also consume Uranium at a relatively high rate, but it sounds like that's only during thrusting.

EDIT: Just to follow up, I'm using KA v1.1.2 (current) in KSP v1.8.1, and the baseline LV-N does not have a generator, but does have an alternator.  Conversely, the Eel (my favorite) and Neptune both have generators that run whether the integrated radiator is turned on or not.  Build a craft with a partially charged battery and an Octo probe core as the only power consumer, and you'll see the battery start to refill.  There's no indicator on the PAW though, other than "Generator: Nominal".   What's interesting is they do have the alternator module, but they're set to a negative output vs the generator output, so thrusting doesn't produce extra power.

I also didn't see anything about Uranium being consumed in the engines I looked at.  So either it's certain engines (I cut out engines bigger than 1.25m to save RAM) or another mod that adds some more realism to consume it in NTRs.

EDIT 2: Also keep in mind, LH2 boils off and CryoTanks adds a power consuming cooling capability (proportional to amount of propellant in the tank) to prevent that.  I think cooling is set to "on" by default in the dedicated Hydrogen tanks, but set to "off" on all the other tanks.  It can be switched on and off in all the tanks.

Edited June 7, 2020 by KSPrynk

[RyGuy_McFly]

8 hours ago, KSPrynk said:

As far as I know, you don't have to turn on a generator in any of the bi-modal NTRs (although there are usually integrated radiators to turn on), like you would for the dedicated nuclear reactor powerplant parts (which come in the Near Future Electrical mod), but I haven't checked to see if I'm a version behind on KA and whether that requirement was recently added.

Thanks for the answer, however after some testing it seems that, at least for me, the LV-N does in fact have to have its reactor started in order to ignite, and also it does not have an alternator. Haven't unlocked any of the other nukes yet, and I do like surprises, so I'll have to wait and see on the rest. I'm also running the same version as you, in 1.8.1

EDIT: I'm using my Nerv in LF mode, does it act differently in LH2 mode perhaps?

Edited June 8, 2020 by RyGuy_McFly

[Dovahkiin2132]

Why aren't the aerospike engines scaleable?

[Nertea]

On 6/27/2020 at 5:55 AM, Dovahkiin2132 said:

Why aren't the aerospike engines scaleable?

Tweakscale -> tweakscale thread

[Iodyne]

Using the patch for NFElectricalNTRs (and EngineLight), the stock nuclear engine still emits light (although not exhaust effects) when the engine is active but the reactor off. Is there a way to edit the config so the enginelight can grab whether the reactor is active or is that a limitation of EngineLight itself? Obviously not a big deal, can always just have the engine deactivated when not using it, but it is an edge case that exists.

[Nertea]

On 7/24/2020 at 2:26 PM, Iodyne said:

Using the patch for NFElectricalNTRs (and EngineLight), the stock nuclear engine still emits light (although not exhaust effects) when the engine is active but the reactor off. Is there a way to edit the config so the enginelight can grab whether the reactor is active or is that a limitation of EngineLight itself? Obviously not a big deal, can always just have the engine deactivated when not using it, but it is an edge case that exists.

Engine light can't talk to NFE. Well it could, but that would be an action on the engine light developers, not me.

[Iodyne]

@Nertea Figured.

[NightSaber11]

does this work with 1.10.1?(i'm new to ksp mods and to ksp as well)

Edited July 30, 2020 by NightSaber11
explanation of my stupidity

[DStaal]

3 hours ago, NightSaber11 said:

does this work with 1.10.1?(i'm new to ksp mods and to ksp as well)

In general, with newly released versions of KSP it takes a while for mods to update.  You'll see the last confirmed worked version in the title of the thread - in this case, 1.9.x.  Anything past that is a question mark.

Pure part mods typically work between KSP versions - though 1.10 was an exception for a lot of mods, as they messed with a couple of things in parts, apparently.  (Notably, no modded fairings worked.)

However, this isn't a pure parts mod - there's some compiled code as well.  Those are more restrictive, and while they often work between versions, they may not.

So, all that said: I haven't tried this mod in 1.10.1, and it may work - but I wouldn't count on it.  Feel free to try it yourself and report if it does, but if it doesn't don't bug the mod-maker about it. 

[Nertea]

Kerbal Atomics 1.1.4

• Marked for KSP 1.10.x
• Updated DynamicBatteryStorage to 2.1.7
• Updated B9PartSwitch to 2.17.0
• Updated CryoTanks to 1.5.2
• Updated DeployableEngines to 1.2.2
• Updated ModuleManager to 4.1.4
• New Chinese localization by tinygrox
• Restock+ Cherenkov now gets a LH2/LF mode switch + custom effects

[Sesshaku]

Quick question, I'm stacking ABSURD amounts of fuel for one of the engines of this mod, and yet I barely get 2-3K of delta V accordin to KER

Am I doing something wrong? Or is it a problem in KER accounting?

Otherwise it seems the consumption of fuel per second is just too big.

[juanml82]

30 minutes ago, Sesshaku said:

Quick question, I'm stacking ABSURD amounts of fuel for one of the engines of this mod, and yet I barely get 2-3K of delta V accordin to KER

Am I doing something wrong? Or is it a problem in KER accounting?

Otherwise it seems the consumption of fuel per second is just too big.

Are you switching the tanks to LH2 only or are they still loaded with LH2 + oxidizer?

[Sesshaku]

2 hours ago, juanml82 said:

Are you switching the tanks to LH2 only or are they still loaded with LH2 + oxidizer?

Yes I did make sure both the engines and the fuel are of the same type. The real problem I'm having is this. Between the massive fuel consumption and the "evaporate" mechanic, I can't seem to actually make an interstellar planet that "works". Let me rephrase it. I can make it work. But it's just too big a ship and the game starts to perform annoyingly.

Before the usual "you're designing the ship with too many piece" keep in mind, I'm using kerbalism, so I have to account for food, oxygen, waste, nitrogen, water, living space, radiation, redundancy, etc. In other words. I was hoping for a mod that allows me to just put 1-2 of the huge spheres of LH2 and be done with it, but no, If I want to reach Jool and come back I have to use like A LOT more than that. And the total piece end up being above 500.

Here are some illustrations of my situation. Now, the pics may look fine, but the performance ain't great.

Photo 1 With just the fuel part. Notice I ended up "cheating". By overpiling fuel in unrealistic ways, there are also "hidden" pieces inside the pieces.

Photo 2 Once I have the complete ship. I am having the following troubles (but this is not the mods fault I always have trouble with this): the amount of Electric charge I have (around 65-85K) depletes for some reason during warping. Which ends up "evaporating" the fuel, which ends up reducing even more the efficiency of the engine. There's an extra that allows you to use LF with the engines, but is there a way of maybe just reducing the LH2 consumption or eliminating the "evaporation" mechanic. I think that with a combination of those two, and a huge rework of the overall ship I could substantially reduce the pieces amount.

Is either that, or just accepting to drastically reduce Kerbalism mechanics, specially the m2 stress requirements, which is the one that adds the more complexity. Still, I really hope the developers make a far better engine for Kerbal Space Program 2, Unity doesn't seem to be capable of dealing with more complex ships.

Edited October 8, 2020 by Sesshaku

[hemeac]

@Sesshaku, I think the issue with EC is that it does get buggy at high timewarp.  Not sure that is a Unity issue, I suspect that the KSP team just didn't prioritize the mechanic at high time warp as they have no life support system and nothing that critically requiring EC at all times.  Back on topic, if the boil-off mechanic causing issues, this untested patch should disable it from all tanks (back up your save first):

// Remove the module that manages the boil-off mechanic from any part that has one
@PART[*]:HAS[@MODULE[ModuleCryoTank]]:FINAL
{
	!MODULE[ModuleCryoTank] {}
}

Alternatively, you can try modifying the boil-off rate and/or cooling cost:

// Alternatively, choose a boil-off rate and cooling cost that fits your style better
@PART[*]:HAS[@MODULE[ModuleCryoTank]]:FINAL
{
	@MODULE[ModuleCryoTank]
	{
		@BOILOFFCONFIG:HAS[#FuelName[LqdHydrogen]]
		{
			// in % per hr
			@BoiloffRate = 0.05 // Change these to a value that you prefer
			@CoolingCost = 0.09
		}
		@BOILOFFCONFIG:HAS[#FuelName[LqdMethane]]
		{
			// in % per hr
			@BoiloffRate = 0.005
			@CoolingCost = 0.045
		}
	}
}

 

[Sesshaku]

8 hours ago, hemeac said:

@Sesshaku, I think the issue with EC is that it does get buggy at high timewarp.  Not sure that is a Unity issue, I suspect that the KSP team just didn't prioritize the mechanic at high time warp as they have no life support system and nothing that critically requiring EC at all times.  Back on topic, if the boil-off mechanic causing issues, this untested patch should disable it from all tanks (back up your save first):

// Remove the module that manages the boil-off mechanic from any part that has one
@PART[*]:HAS[@MODULE[ModuleCryoTank]]:FINAL
{
	!MODULE[ModuleCryoTank] {}
}

Alternatively, you can try modifying the boil-off rate and/or cooling cost:

// Alternatively, choose a boil-off rate and cooling cost that fits your style better
@PART[*]:HAS[@MODULE[ModuleCryoTank]]:FINAL
{
	@MODULE[ModuleCryoTank]
	{
		@BOILOFFCONFIG:HAS[#FuelName[LqdHydrogen]]
		{
			// in % per hr
			@BoiloffRate = 0.05 // Change these to a value that you prefer
			@CoolingCost = 0.09
		}
		@BOILOFFCONFIG:HAS[#FuelName[LqdMethane]]
		{
			// in % per hr
			@BoiloffRate = 0.005
			@CoolingCost = 0.045
		}
	}
}

 

Oh, that's neat. Thank you for your quick answers. Unfortunately, not only I'm a nuisance, I'm also not a programmer. So quick question though, how do I apply the patch? Do I add the @PART bit somewhere in cryotanks folder? Also, the part of fuel consumption, is it decided on the "ratio" part that I found here?

PROPELLANT
        {
            name = LqdHydrogen
            ratio = 1.0
            DrawGauge = True
        }

Without evaporation and with half the consumption I'm certain I can reduce at least 30 parts off my ship. I also think I'm gonna reduce the m2 per kerbal required on kerbalism, and likely making it "more tight" in terms of food,water and oxygen. Which removes some realism, but at least won't make me feel like I'm playing on a potato.

 

 

[juanml82]

12 hours ago, Sesshaku said:

Yes I did make sure both the engines and the fuel are of the same type. The real problem I'm having is this. Between the massive fuel consumption and the "evaporate" mechanic, I can't seem to actually make an interstellar planet that "works". Let me rephrase it. I can make it work. But it's just too big a ship and the game starts to perform annoyingly.

Before the usual "you're designing the ship with too many piece" keep in mind, I'm using kerbalism, so I have to account for food, oxygen, waste, nitrogen, water, living space, radiation, redundancy, etc. In other words. I was hoping for a mod that allows me to just put 1-2 of the huge spheres of LH2 and be done with it, but no, If I want to reach Jool and come back I have to use like A LOT more than that. And the total piece end up being above 500.

Here are some illustrations of my situation. Now, the pics may look fine, but the performance ain't great.

Photo 1 With just the fuel part. Notice I ended up "cheating". By overpiling fuel in unrealistic ways, there are also "hidden" pieces inside the pieces.

Photo 2 Once I have the complete ship. I am having the following troubles (but this is not the mods fault I always have trouble with this): the amount of Electric charge I have (around 65-85K) depletes for some reason during warping. Which ends up "evaporating" the fuel, which ends up reducing even more the efficiency of the engine. There's an extra that allows you to use LF with the engines, but is there a way of maybe just reducing the LH2 consumption or eliminating the "evaporation" mechanic. I think that with a combination of those two, and a huge rework of the overall ship I could substantially reduce the pieces amount.

Is either that, or just accepting to drastically reduce Kerbalism mechanics, specially the m2 stress requirements, which is the one that adds the more complexity. Still, I really hope the developers make a far better engine for Kerbal Space Program 2, Unity doesn't seem to be capable of dealing with more complex ships.

I don't use Kerbalism, so I don't know how it controls for electricity usage while the ship in "on rails" mainly because there is no "on rails" with kerbalism. But what I do is to use nuclear reactors (from NF electric) to power the tanks. Since the (stock) game computes electricity usage at the time you jump away from the ship, after I finish burning towards destination I set up a KAC alarm for either the SOI change or a maneuver node, make sure the ship isn't loosing power and then I go back to the space center (or any other ship) rather than warping all the way to Jool because KSP doesn't handle electrical production during high time warp well.

Since the trip to Jool is long, you may need to carry additional fuel for the nuclear reactors, as those tanks are power hungry. I'm at work, so I don't have a screenshot at hand, but one way I design ships to minimize power requirements is to build them around 5m tanks rather than the large spherical ones and then radially attach the smaller spherical tanks in rings alongside the central tank. That way, you can set the fuel priorities of the external spherical tanks so they are used from bottom to top and stage them as they become depleted, so you no longer need to power them. Essentially: the payload sits atop two 5m cryotanks with the engine at the bottom. Around those tanks, there are four "rings" of six radial decouplers (per "ring"), each with fuel crossfeed enabled and holding a smaller spherical tank.

That means you'll need less power while coasting towards Jool because one or two rings would have been staged while making the trans-Jool injection burn at Kerbin. On the downside, if you're refueling at your destination for the trip back, you won't have the ship's full fuel capacity, as you'll only have the center tanks for the journey back.

[Brigadier]

5 hours ago, Sesshaku said:

how do I apply the patch?

@Sesshakusimply copy the MM code into a txt file, and save it with a descriptive name and a .cfg extension.  Place the file in your GameData folder (I use a unique folder within GameData to store all of these personal .cfg files so they're easy to find and aren't overwritten by mod updates).  MM reads all .cfg files in GameData at loadup.  Since these two MM patches are marked FINAL, they'll be implemented towards the end of the MM patch sequence.

[Nertea]

15 hours ago, Sesshaku said:

Photo 2 Once I have the complete ship. I am having the following troubles (but this is not the mods fault I always have trouble with this): the amount of Electric charge I have (around 65-85K) depletes for some reason during warping. Which ends up "evaporating" the fuel, which ends up reducing even more the efficiency of the engine. There's an extra that allows you to use LF with the engines, but is there a way of maybe just reducing the LH2 consumption or eliminating the "evaporation" mechanic. I think that with a combination of those two, and a huge rework of the overall ship I could substantially reduce the pieces amount.

Yes, this is a problem with stock KSP. This mod bundles another mod that attempts to fix this, but it is disabled in the presence of Kerbalism because they do their own thing.